Contact name: 
Paul Willner
Centre for Substance Abuse Research, Department of Psychology, University of Wales, Swansea
+44 (0) 1792 295844
Centre for Substance Abuse Research, University of Wales Swansea, SwanseaSA2 8PP, UK.
( + 44) 1792 295679



A number of studies have reported high levels of alcohol consumption among adolescents in the United Kingdom. While little is known about variables that might be contributing to these high rates of teenage alcohol use, ease of access to alcohol may be a major factor (Wagenaar et al., 1993, 1996). It appears that many commercial alcohol outlets fail to comply with minimum age laws, and that alcohol is readily available, from a variety of different types of outlets, to underage adolescents who wish to purchase it.


The aims of this study were to assess the ease with which adolescents in the United Kingdom are able to buy alcohol, to obtain information concerning vendors perceptions of alcohol sales to adolescents, and to evaluate a police intervention intended to reduce underage alcohol sales.


An unobtrusive naturalistic field study was conducted in two urban locations. Pairs of 13- and 16-year-old boys and girls were trained to attempt the purchase of different types of alcohol (alcopops, beer, cider, wine, spirits) from four different types of retail outlets (corner shops, off-licence, public houses and supermarkets), under the supervision of a researcher and typically a parent. The assessment was repeated, with the omission of the 13-year-old boys, following a police intervention in one of the performance sites, consisting of warning letters and visits to vendors, and the issue of a small number of police cautions. A total of 62 underage confederates in all attempted 470 test purchases in phase 1 and 348 in phase 2. Between the two waves of test purchases a sample (n=95) of the same vendors was surveyed by telephone.


In phase 1, sales resulted from 88.1% of purchase attempts by 16-year-old girls, 77% of attempts by 16-year-old boys, 41.6% of 13-year-old girls and 4.1% of 13-year-old boys. These figures were generally comparable across locations, alcohol types and outlet types. Refusals were more likely when another vendor was present. Eighty per cent of sales to 16-year-olds and 65% of sales to 13-year-old girls were made without challenge. Prove-It ID cards were requested in fewer than 12% of purchase attempts in both age groups. Overall, there was no evidence that the police intervention reduced sales of alcohol to 16-year-olds. There was a hint that the intervention may have caused a very short-lasting decrease in sales to 13-year-old girls, but this was contained within an overall increase in sales to this group. Alcohol vendors reported that they rarely encountered underage customers or refused sale although 90% of vendors said that if they became suspicious, they would request ID. Only two vendors believed that they were likely to suffer adverse consequences if they sold alcohol to minors.


These data suggest that 16-year-olds, and girls as young as 13, have little difficulty in purchasing alcohol, and that there is little difference between different types of outlets in their willingness to sell alcohol to minors. Vendors perceive little risk in selling alcohol to adolescents. The fact that the police intervention failed to decrease sales suggests that vendors do not change their behaviour in response to the threat of legal action.

Intervention details

Type of intervention
Other enforcement
Problem addressed
Alcohol, Underage drinking, Legislation
Intervention setting
Corner shops
Target population


Substances adressed
Strategic target group (social agents acting as intermediaries between intervention and target group)

Vendors of alcoholic beverages

Intervention activities
Nightlife management
The police intervention consisted of the following elements:A letter from the area police commander was sent to all licenced premises within the test areas:reporting the headline figures from our first round of test purchasing;reminding licencees of their legal obligations with respect to the sale of alcohol to minors;recommending that proof of age should be sought from any young person who appeared to be below the age of 21;recommending that only a passport or the Prove-It card issued by the Portman Group should be accepted as proof of age;indicating that the police would shortly be instituting a campaign in which they would themselves carry out test purchases for evidentiary purposes.The contents of the letter were reiterated in personal visits to most licenced premises and telephone calls to regional and area managers of the major brewery, off-licence and supermarket chains;A press release was issued, which resulted in extensive coverage of the initiative by the local press and local radio;A total of 11 test purchases were subsequently carried out by a 13-year-old boy who looked no more than 14 and did nothing to deceive licencees as to his age. Three of these test purchases resulted in a sale. All three offenders were warned that any further offence could result in prosecution, and this was also publicized in the local media.
Theory/evidence behind the intervention

There have been some American studies in which young-looking 21-year-olds (21 being the legal age to purchase alcohol in the United States), were recruited to make test purchases of alcohol (Wagenaar et al., 1993; Forster et al., 1994): for example, Forster et al. (1994) found that 47% of a sample of 21-year-olds judged to be 19 or younger were able to buy beer. Vaucher et al. (1995), working in Switzerland, where beer may be bought at 16 and pastis (an aniseed-flavoured aperitif) at 18, found that both drinks were sold to 13- and 15-year-old boys on 81% of purchase attempts. The original aim of the present study was to apply the methodology of Vaucher et al. (1995) to the United Kingdom.

Evaluation details

Evaluation type (e.g. process, outcome, cost-effectiveness)
Outcome evaluation
Activities evaluated

Test purchases and police intervention

Evaluation results (Outcome evaluation)

In the first wave of purchase attempts, prior to the intervention, sales to 16-year-old girls were slightly lower in the control site than in the intervention site, but the two groups of 16-year-old boys and 13-year-old girls were very well matched. In phase 2, sales to 16-year old girls in the control site increased to a level comparable to that in the intervention site, while sales to 16-year-old boys and 13-year-old girls did not change significantly.

Evaluation references

PAUL WILLNER, KENNETH HART, JOHN BINMORE, MARGARET CAVENDISH & ELIZABETH DUNPHY (2000). Alcohol sales to underage adolescents: an unobtrusive observational field study and evaluation of a police intervention. Addiction, 95, 9, 1373-1388.

An abstract for this journal article can be found in the HNT literature section here.